A comparative study of moment-based shape descriptors for product image retrieval

Author

Kejia, Wang ; Honggang, Zhang ; Lunshao, Chai ; Haiying ; Ziliang Ping

Author_Institution

Sch. of Electron. Eng., Beijing Univ. of Posts & Telecommun., Beijing, China

fYear

2011

fDate

21-23 Oct. 2011

Firstpage

355

Lastpage

359

Abstract

This paper presents a comparative study of Zernike moments (ZMs), Jacobi-Fourier moments (JFMs) and Radial-Harmonic-Fourier moments (RHFMs) for product image retrieval. Experiments are carried out on the PI 100 dataset and the retrieval performances of the shape descriptors are evaluated and quantified by the Average Normalized Modified Retrieval Rank (ANMRR). The experimental results show that JFMs and RHFMs present similar better retrieval performance than ZMs in both terms of shape representation quality and the amount of required moments.

Keywords

Zernike polynomials; image representation; image retrieval; method of moments; ANMRR; JFM; Jacobi-Fourier moments; PI 100 dataset; RHFM; Radial-Harmonic-Fourier moments; ZM; Zernike moments; average normalized modified retrieval rank; moment-based shape descriptors; product image retrieval; required moments; retrieval performances; shape representation quality; Harmonic analysis; Image edge detection; Jacobi-Fourier moments (JFMs); Radial-Harmonic-Fourier moments (RHFMs); Zernike moments (ZMs); image retrieval; shape descriptor;

fLanguage

English

Publisher

ieee

Conference_Titel

Image Analysis and Signal Processing (IASP), 2011 International Conference on

Conference_Location

Hubei

Print_ISBN

978-1-61284-879-2

Type

conf

DOI

10.1109/IASP.2011.6109062

Filename

6109062